Baseline score
A baseline score is the committed minimum evaluation score an agent version must meet, expressed between 0 and 1. Baselines are monotonic — they may only increase across versions — so an agent's measured quality can never silently regress as prompts, tools or models change.
In NeoXam Agents
Each shipped agent carries an eval baseline in the 0.70–0.80 range, versioned in the Catalog alongside its eval dataset. Raising the baseline is part of promoting a new agent version; lowering it is not an option the platform offers.
Related terms
See how these concepts come together in a governed agentic platform for investment operations.
Explore the platform